Associations between pulsed oxygen saturation during out-of-hospital cardiopulmonary resuscitation, cardiopulmonary resuscitation quality and return of spontaneous circulation.
Resuscitation · 2026-07-23 · Observational study
Abstract
BACKGROUND: The association between pulsed oxygen saturation (SpO2) and out-of-hospital cardiac arrest (OHCA) survival outcomes is unknown. We hypothesized that higher event-average SpO2 is associated with higher probability of return-of-spontaneous-circulation (ROSC). METHODS: We analyzed the association of the event-average SpO2, defined as the average of all recorded values during the resuscitation event, with sustained ROSC. Secondary exposures included survival to hospital admission, favorable neurological outcome (Cerebral Performance Category <3 or no change from baseline) and associations between CPR quality metrics and cycle-average SpO2, defined as the mean SpO2 value within each two-minute CPR cycle. RESULTS: We included 98 index OHCA with mean age of 62.1 years, 32% female, 9.2% shockable rhythm. Median event-average SpO2 was 55.0 ([Q1, Q3], 38.8, 72.3). Events with sustained ROSC had a higher event-level average SpO2 (72.3 [57.0, 82.9] versus 48.0 [34.9, 65.4]; P < 0.001). The receiver operating characteristic analyses of SpO2, as a predictor of sustained ROSC, identified an area under the curve of 0.77 [95% CI, 0.68-0.86]) with an optimal cut point at 71.0 (sensitivity 58, specificity 0.85). After adjusting for confounders, SpO2 was associated with higher probability of sustained ROSC (adjusted odds ratio, 1.05 [95% CI, 1.02-1.08]; P < 0.001). Event-average SpO2 was also associated with higher probability of survival to hospital admission (adjusted odds ratio, 1.03 [95% CI, 1.01-1.05]; P = 0.005) while it was not associated with higher probability of neurological intact survival (adjusted odds ratio, 1.00 [95% CI, 0.96-1.05]; P = 0.79). Higher cycle-average SpO2 was not associated with better chest compressions metrics. CONCLUSIONS: Higher event-average SpO2 was associated with higher probability of unadjusted and adjusted sustained ROSC but not with neurological intact survival. Cycle-average SpO2 was not associated with chest compressions quality metrics.